Keywords
epistaxis - children - outpatient clinic - bleeding disorder
Introduction
Epistaxis is defined as acute bleeding from the nostril, nasal cavity, nasopharynx,[1]
[2] or the paranasal sinuses,[3] and it is a common problem among children, representing a prevalent emergency in
otolaryngology. Most nosebleeds are benign, self-limited, and rarely life-threatening.[4] It is crucial to diagnose the cause of the bleeding, whether it is local or secondary
to inherited systemic coagulopathy, as this strongly affects the management and might
require the presence of a pediatric hematologist in the management team, alongside
the otolaryngologist.[5]
There is limited data regarding the prevalence of nosebleeds in children because most
episodes are self-limited and, thus, not reported. When medical attention is needed,
it is usually because of either the recurrent or severe nature of the problem.[6] It was estimated previously that 60% of the general population will have an episode
of epistaxis at some point in their lives.[7] On the other hand, epistaxis is rare in children younger than 2 years of age (∼
1 per 10,000) and should raise prompt consideration of trauma (intentional or unintentional),
serious illness (e.g., thrombocytopenia),[8] or child abuse.[9]
Although most of the severe cases of inherited bleeding disorders will be diagnosed
in an early stage of life, moderate and mild cases may not be detected.[10] We believe that in our region, the prevalence of bleeding disorders is higher than
the international prevalence, due to a higher rate of marriages among family relatives;
however, the exact rate is unknown. The current cross-sectional study aims to measure
the frequency and etiology of epistaxis as a presenting symptom among children who
visited a tertiary pediatric outpatient clinic and to determine the frequency of secondary
epistaxis due to bleeding tendency using the available simple assessment tools (bleeding
score and epistaxis severity score [ESS]). Our hypothesis was that the use of a simple
tool of assessment could guide in subgrouping those patients with epistaxis who might
need further assessment. This could increase the diagnosis rate of inherited bleeding
disorders in children presenting with epistaxis and, eventually, improve the care
given to the affected children.
Patients and Methods
This is a cross-sectional study in which the subjects were recruited from the outpatient
clinic of the Pediatrics Hospital of the Ain Shams University over 6 working days
(from Saturday to Thursday), in the working hours from 9 am until 12 pm, over a 1-year study period starting in December 2016.
The participants were pediatric patients (younger than 18 years) who presented to
the outpatient clinic with epistaxis of unknown cause. Patients older than 18 years
or having epistaxis due to evident surgical or traumatic cause or because of an evident
medical cause, such as an inherited bleeding disorder, liver-cell failure, or because
of administration of aspirin or warfarin, were excluded.
An informed consent was obtained from the subjects and/or their legal guardians before
participation. The procedures applied in this study were approved by the Ethical Committee
of Human Experimentation of the Ain Shams University and are in accordance with the
Helsinki Declaration of 1975 with Federal wide assurance (FWA) number: 000017585.
All the recruited patients underwent thorough clinical assessment with special emphasis
on the demographic data, history of trauma or any chronic illness, history of current
infection or inflammation, history of other bleeding manifestations, and family history
of epistaxis, bleeding tendency or blood product transfusion. An ear, nose, and throat
(ENT) examination was performed by an assigned trained physician using anterior rhinoscopy.
The following two simple questionnaires were completed:
-
The pediatric bleeding questionnaire (PBQ):[11] The online version of this questionnaire can be accessed in the “Supporting Information”
part of the article's citation (https://onlinelibrary.wiley.com/doi/full/10.1111/j.1538-7836.2010.03975.x/),
-
Epistaxis severity score (ESS): Self-reported questionnaire on the frequency of epistaxis,
duration and amount of attack, proportion of child's life that the nosebleed has been
recurrent, and whether the nosebleed is unilateral or bilateral. Each component was
scored 0, 1, or 2. The sum of the scores of all components was stratified as mild,
if it was 0 ≥ 6, or severe if 7 ≥ 10.[12]
The patients were classified into two groups according to whether their bleeding was
only restricted to the nose (group A: Patients with probable local bleeding) or associated
with other bleeding manifestations (group B: Patients with probable systemic bleeding).
Peripheral blood samples were collected under complete aseptic conditions for the
evaluation of complete blood count (CBC), prothrombin time (PT), and partial thromboplastin
time (PTT). Two milliliters of fresh peripheral blood samples were collected on potassium
ethylene diamine tetra-acetic acid (K2-EDTA) (1.2 mg mL − 1) as an anticoagulant to
perform CBC on Sysmex XT-1800i (Sysmex, Kobe, Japan). For PT and PTT, two milliliters
of fresh peripheral blood samples were obtained on sodium citrated vacutainer (1-part
citrated sodium (0.11 mol/L) with nine parts venous blood, (platelet poor plasma)
was collected by centrifugation at 3,000 × g for 15 minutes and was performed on a
Stago autoanalyzer Dubai (UAE).
Confirmatory investigations were performed in patients who showed abnormalities in
their basic bleeding profile (CBC, PT, and PTT) as well as in those with normal basic
bleeding profile yet a bleeding score exceeding ≥ 2, which may indicate an underlying
bleeding tendency. Confirmatory investigation was performed according to abnormality
detected and included platelet functions analysis, coagulation factor assay, and Von
Willebrand factor (VWF) antigen and cofactor activity assessment.
Statistical Analysis
The data were collected, revised, coded and entered to the IBM SPSS Statistics for
Windows, Version 23.0 (IBM Corp., Armonk, NY, USA). Quantitative variables were described
in the form of mean and standard deviation or median and interquartile range (IQR;
75th and 25th percentiles). Qualitative variables were described as number and percentages.
To compare parametric quantitative variables between two groups, the Student t-test was applied. For comparison of non-parametric quantitative variables between
two groups, the Mann-Whitney test was used. Qualitative variables were compared using
the chi-squared (χ2) test or Fischer exact test when frequencies were below five.
Receiver operating characteristic (ROC) curves were used to assess the best cut-off
point with sensitivity, specificity, positive, and negative predictive value (PPV
and NPV) and area under the curve (AUC). The confidence interval was set to 95%, and
the margin of error accepted was set to 5%. So, the p-value significance was set as follows: P-value > 0.05 was considered non-significant (NS), and P-value < 0.05 was considered significant in all analyses.
Results
A total of 30,043 patients presented to the outpatient clinic of the Ain Shams University
Hospital through the 1-year study period; 106 patients presented with epistaxis, and
6 of them were lost to follow-up or had incomplete data; only 100 patients with complete
data were enrolled in the study, with an estimated frequency of epistaxis of around
1 in 300 annually. Out of the 100 recruited patients, 53 were female and 47 were male,
and all of them were older than 2 and younger than 18 years old, with a mean of 7.89 ± 3.89
years. Most of the recruited patients were younger than 12, and nearly half of them
were younger than 6 years old; 48% were aged 2 to 6 years, 36% were 6 to 12, and 16%
were 12 to 18 years. In our study, 29 cases had consanguineous parents and 71 cases
did not, and nearly half of the recruited patients (54%) had positive family history
of epistaxis. Seventy-six patients had recurrent attacks of epistaxis, and 24 patients
had a single attack.
Seventeen out of 100 recruited patients had comorbidities. Five of them had local
comorbidities: three patients had allergic rhinitis, one had paranasal sinusitis,
and one had concha bullosa. The other 12 patients had systemic comorbidities, mainly
diabetes mellitus (4 patients) and gastrointestinal disease (3 patients). There was
a significantly higher percentage of patients who presented with epistaxis than those
who presented with other complaints in the warm months of June, July, August, and
September, and a lower percentage in the spring months of February and March, as illustrated
in [Fig. (1)]. Most of the patients (90%) presented with anterior bleeding; 10 had undetermined
site of bleeding, and none had posterior bleeding. One hundred patients presented
with epistaxis; most of them (96%) were treated with only nasal compression as a procedure,
while 4 had anterior pack and 2 underwent cauterization.
Fig. 1 The percentage of patients visiting the outpatient clinical of Ain Shams University
Pediatric hospital in compare to the percentage of patients recruited in our study
(those presented with epistaxis) monthly.
Forty-three patients presented with epistaxis only (group A: patients with suspected
local bleeding); 37 of them had normal ENT assessment and normal initial hematological
assessment (CBC, PT, and PTT), as illustrated in [Table 1]. They were diagnosed with idiopathic epistaxis; 6 patients had local ENT causes
(4 had allergic rhinitis and 2 had infections).
Table 1
Comparison between local and systemic cause of epistaxis in regards to the studied
parameters
|
Group B: Patients with probably systemic bleeding N = 57
|
Group A: Patients with probably local bleeding
N = 43
|
Test value
|
P-value
|
|
Sex; N (%)
|
Female
|
30 (52.6%)
|
23 (53.5%)
|
0.007[*]
|
0.932
|
|
Male
|
27 (47.4%)
|
20 (46.5%)
|
|
Age; N (%)
|
2–6 years
|
16 (37.2%)
|
32 (56.1%)
|
3.763[*]
|
0.152
|
|
> 6–12 years
|
19 (44.2%)
|
16 (28.1%)
|
|
> 12–18 years
|
8 (18.6%)
|
9 (15.8%)
|
|
Consanguinity; N (%)
|
19 (33.3%)
|
10 (23.3%)
|
1.209[*]
|
0.272
|
|
Family history; N (%)
|
25 (43.9%)
|
29 (67.4%)
|
5.487[*]
|
0.019
|
|
Recurrent; N (%)
|
42 (73.7%)
|
34 (79.1%)
|
0.390[*]
|
0.532
|
|
Seasonal variation; N (%)
|
18 (31.6%)
|
23 (53.5%)
|
4.864[*]
|
0.027
|
|
Referred patients; N (%)
|
27 (47.4%)
|
4 (9.3%)
|
16.604[*]
|
0.000
|
|
Total leucocytic count (103/uL); mean ± SD (range)
|
7.75 ± 2.19
(4–12.1)
|
7.71 ± 2.66
(4–16.7)
|
0.087[•]
|
0.931
|
|
Hemoglobin (gm/dl); mean ± SD (range)
|
11.19 ± 1.56
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
|
11.14 ± 0.81
[9]
[10]
[11]
[12]
[13]
|
0.212[•]
|
0.833
|
|
Platelets (103/uL); mean ± SD (range)
|
202.68 ± 140.13
(4–500)
|
258.72 ± 85.00
(127–455)
|
-2.319[•]
|
0.023
|
|
Pediatric bleeding score; median (IQR)
|
6 (4–9)
0–20
|
1 (0–1)
0–7
|
-7.458ǂ
|
0.000
|
|
Epistaxis severity score; median (IQR)
|
4 (2–6)
0–10
|
2 (0–5)
0–8
|
-2.098ǂ
|
0.036
|
* Chi-squared test
• Independent t-test.
Qualitative data are presented as number and percentages, in which the chi-squared
test was used for comparisons. Quantitative data are presented as mean with standard
deviations, when parametric, in which the independent t-test was used, and as median with interquartile ranges (IQRs), when nonparametric,
in which the Mann-Whitney tests was used.
Fifty-seven patients presented with other bleeding manifestations (group B: patients
with suspected systemic bleeding). The most frequent associated bleeding manifestations
were cutaneous bleeding; petechiae (32 patients) and ecchymosis (25 patients), followed
by bleeding gum (18 patients) and menorrhagia (14 patients). The least frequent associations
were bleeding per rectum (4 patients) and bleeding after tooth extraction (3 patients).
The pediatric bleeding score and ESS of this group are illustrated in [Table 1]. Patients with suspected systemic bleeding and/or those who had abnormal initial
hemogram and coagulation profile performed further laboratory tests to reach a final
diagnosis. The VWF antigen assay and VW cofactor activity as well as platelet function
tests with adenosine diphosphate and Ristocetin were performed for 32 patients; consequently,
12 patients had platelet function defect (PFD) and 10 patients had Von Willebrand
disease (VWD). Only 6 patients needed coagulation factor assay and revealed that 1
patient had factor-7 deficiency, 1 patient had hemophilia A and one patient had hemophilia
B. Only eight patients underwent bone marrow (BM) examination; they had bipancytopenia
or associated hepatosplenomegaly and generalized lymphadenopathy; four had abnormal
BM aspiration (BMA); one case had myelodysplasia, and three cases had aplastic anemia.
Four patients had normal BMA with peripheral platelet destruction and were diagnosed
as idiopathic thrombocytopenic purpura (ITP) .
Consequently, 47 patients had a definite bleeding disorder and 10 had undiagnosed
bleeding tendency; 16 had immune mediated disease (14 patients had ITP and 2 patients
had autoimmune diseases [antiphospholipid antibody syndrome and systemic lupus], 25
had inherited defect [12 patients had PFD, 10 patients had VWD, one had factor 7 deficiency,
one had hemophilia A and one had hemophilia B]and 6 had other diseases (one case had
myelodysplasia and 3 cases had aplastic anemia, one patient had DIC, one patient had
poststreptococcal glomerulonephritis and hypertension). Out Of them 43 patients are
now regularly following up at pediatric hematology clinic as shown in [Fig. (2)].
Fig. 2 Schematic approach of screened and recruited patients.
We compared patients presenting with epistaxis only (local bleeding) with those presenting
with other types of bleeding to identify the predictors of bleeding tendency in patients
presenting with epistaxis, as illustrated in [Table 1]. There was a statistically significantly higher number of patients with a positive
family history of epistaxis in those with suspected systemic cause. Low platelets
count significantly alerts for probably a systemic cause of epistaxis. The presence
of seasonal variation of epistaxis was statistically significant in favor of the local
cause of epistaxis. There was a statistically higher number of patients with suspected
systemic cause of epistaxis, and these were referred from a primary care physician
to our outpatient clinic.
The ROC curve analysis showed that a bleeding score of 2 or more was of a high statistical
significance in the detection of an underlying systemic cause of bleeding, as it gave
sensitivity of 85.96% and specificity of 95.35% with the area under the curve (AUC)
being 0.932, PPV of 96.1, and NPV) of 83.7. Whereas ESS of 2 or more showed statistically
significant value in the detection of an underlying systemic cause of epistaxis, as
it gave sensitivity of 66.67% and specificity of 58.14% with the AUC being 0.621,
PPV of 67.9, and NPV of 56.8. A boxplot ([Fig. 3]) showed that bleeding score and ESS were statistically significantly higher in those
with possible systemic underlying causes of epistaxis than those with possible local
causes.
Fig. 3 The values of bleeding score (2A) and epistaxis severity score (2B) in differentiated
between local and systemic underlying cause of epistaxis.
Discussion
Epistaxis is one of the main complaints in ENT outpatient clinics[13]
[14]; it is considered the second most common cause of emergency admission following
sore throat.[13] Our estimated frequency of epistaxis in a pediatric outpatient clinic setting, with
exclusion of traumatic causes, was 1 in 300. This frequency was not far from a 10-year
study, which found that epistaxis accounts for ∼ 4 in 1,000 pediatric emergency department
visits among children in the United States.[7]
None of our recruited patients was younger than 2 years, which is in agreement with
the fact that epistaxis rarely occurs in children under 2 years of age.[15]
[16] The majority of our recruited patients were younger than 12 and adolescents had
the lowest frequency, which agrees with a Brazilian study that reported that up to
60% of children experience one episode of epistaxis prior to the age of 10.[17] Furthermore, Haddad[16] reported that the commonest age for its occurrence is 3 to 8 years old, yet, this
is a little bit lower in the ages of 10 to 15, and, finally, the incidence of epistaxis
decreases after puberty.
The gender distribution in our study showed nearly equal frequency in both genders;
this agrees with the results obtained in a Turkish study,[18] which showed that 52.5% patients were male and 47.5% were female. However, it contrasts
with the findings of Hassan et al.,[19] who conducted a study in Bangladesh in which the majority (68%) of his patients
was male, with the male to female ratio being 2:1; the higher percentage of males
was explained by the higher incidence of traumatic cause in the enrolled patients.
In family history analysis, 54% of the patients had positive family history, and these
had a higher incidence of suspected systemic cause of epistaxis than those with negative
family history. This could be explained by the high rate of consanguineous marriages
(29%) among the studied patients and the relatively high rate of rare factor deficiencies
due to consanguinity of the parents.[18]
Seasonal variations in the occurrence of epistaxis have been noted, but studies regarding
the association of epistaxis and climatic events have yielded inconsistent results.[17]
[20]
[21]
[22] We reported significant monthly variation in recruiting the patients, as there were
a significantly higher number of patients presenting with epistaxis in the warm months
of the year and a lower number during the spring; this variation may be attributed
to seasonal variation. Our results were similar to those of a Turkish study,[23] which found that epistaxis occurrence was positively related to mean temperature.
This finding contradicts the results obtained in a study conducted in the United States
in which epistaxis occurred more often in the fall and winter, when temperatures are
lower.[24]
[25]
[26]
Epistaxis causes can be broadly classified into either local or systemic causes.[27] The local causes may be in the form of inflammation and infection within the nares,
nasal cavity, nasopharynx, and sinuses, dry air exposure,[28] local irritation by chemicals,[29] vascular anomalies,[30] local digital manipulation—the most common cause—[31] and erosion of the nasal mucosa in the Little region of the nasal septum,[32] which is the most common cause in adolescents. But, in most of cases, the etiology
is not determined.[33] Allergic rhinitis has been implicated as a cause of epistaxis episodes in children,
potentially due to direct irritation as a response to associated inflammation, congestion,
and rhinorrhea.[34] In our study, allergic rhinitis was the most common local comorbidity in children
with epistaxis.
Almost all types of bleeding disorders, both congenital and acquired,[35] may present with epistaxis.[36]
[37] In children with severe bleeding disorders, the bleeding history is usually clear;
however, children with mild-to-moderate bleeding disorders may present with recurrent
epistaxis, that is a common complaint in children,[38] and often causes significant anxiety in patients and their family members.[2] Both before and during the diagnosis, these patients may go through considerable
clinical problems and cause diagnostic and therapeutic challenges and an economic
burden.[39]
[40] Screening for bleeding disorders in children with controllable episodes may often
be overlooked,[41] so referral to a specialized hematologist is of value in diagnosing and sparing
these patients the risk of future unexplained bleeding without a clear diagnosis and
management protocol.[42]
Nearly half of the studied patients were at last diagnosed as having a bleeding disorder.
These results are very close to those of a retrospective Turkish study[18] of 26,737 patients of an ambulatory pediatrics division, between 31 October 2011
and 31 October 2012. Ninety-nine patients presented with bleeding symptoms; 36 of
these presented with epistaxis, and nearly 1/3 of them (36.4%) were diagnosed with
a bleeding disorder, after exclusion of all patients with immune thrombocytopenia.
The high incidence of VWD in both studies (12 patients in Turkish study,[18] 10 patients in our study) could be explained by the fact that both studies were
conducted at a tertiary hospital and in Middle Eastern countries, where consanguineous
marriages are common.[43] Furthermore, Egyptian studies conducted at the Pediatric Hematology clinic of the
Ain Shams University revealed that epistaxis represented 15% of the presenting symptoms
of inherited coagulation disorders[43] and were reported in 10% of pediatric ITP.[44] Another Egyptian study showed that 47.6% of patients with recurrent epistaxis had
bleeding disorders[45] and that bruising and epistaxis were the main symptoms in all children with VWD.[46]
Epistaxis can be classified into recurrent and single-episode epistaxis. Recurrent
epistaxis is a common occurrence in children, estimated to affect up to 9% of the
pediatric population[42] and represents a concern for both child and parents. Recurrent epistaxis is even
of more important, as it rises our suspicion of an underlying bleeding disorder. In
patients with recurrent epistaxis, the rate of detecting a bleeding disorder is 5.5
to 33%.[31]
[47] Because emergency providers may often be the first line of evaluation of a patient
with recurrent epistaxis, their understanding of the symptoms and laboratory assessments
leading to a diagnosis of bleeding tendencies is, therefore, critical to identifying
and properly managing these disorders.[31]
[47] Among patients manifesting with recurrent epistaxis, the rate of detection of a
bleeding disorder had a very wide range of 5.5 to 33%.[31]
[47] Most of our studied patients had recurrent attacks of epistaxis (76%); of them,
42 (55%) patients had systemic cause and 34 had local cause.
The treatment of epistaxis requires a systematic and methodical approach, and options
vary according to the cause, location, and severity of the hemorrhage.[48] The majority of our pediatric patients had anterior bleeding, which is consistent
with the data available in the literature,[19]
[31]
[49] and most them (96%) were treated with nasal compression only, which is also consistent
with another study in which the majority (88%) of the patients were left alone with
medical advice and observed.[19]
In general, clinical judgment with consideration of the patient's clinical presentation,
symptoms, age, family history, and degree of bleeding guide the outpatient evaluation
of pediatric epistaxis, and, eventually, most of the epistaxis cases require no laboratory
testing.[50] Unfortunately, the published data do not clearly define which pediatric epistaxis
cases require further laboratory investigation for bleeding disorders.[47] Although patients who present with these symptoms may have underlying bleeding disorders,
initial tests for bleeding etiology may yield normal results. Therefore, other factors
should be considered which may predispose epistaxis in patients and may affect hemostasis.[47]
The first reported application of a bleeding questionnaire was by Rodeghiero et al.[11] Since then, several questionnaires have been evaluated and validated for diagnosing
bleeding disorders in children presenting to the office with bleeding complaints.[51] Our hypothesis was that the use of simple tool of assessment could guide in subgrouping
those patients with epistaxis who in need for further assessment. We found that a
PBQ of 2 or more had high significance in detecting an underlying systemic cause of
bleeding; it gave sensitivity of 85.9% and specificity of 95.35. The sensitivity,
specificity, PPV and NPV of the PBQ were 83%, 79%, 0.14, and 0.99, respectively, in
an earlier study.[10] Consequently, we recommend the use of the PDQ to predict the need to be referred
to a pediatric hematologist for assessment. Furthermore, we found that the referrals
from the primary care physician (PCP) showed higher incidence of having a systemic
underlying cause of epistaxis. This may be due to the fact that non-hematologist PCPs
are less likely to encounter individuals with suspected bleeding disorders, and this
highlights the importance of good communication and referral of non-straightforward
cases to a specialized hematologist to reach a diagnosis and receive a better management,
which spares these patients from the risk of future unexplained bleeding without a
clear diagnosis and management protocol.
Conclusion
The PDQ is a useful and simple tool in the identification of pediatric patients who
present to pediatric outpatient clinics in need of further diagnostic testing to detect
any underlying bleeding tendency.
